Patents

US 12094182 B2: Neural network based identification of areas of interest in digital pathology images

Granted
2024-09-17
Filed
2020-05-29 application 17416394
Type
Utility · 20 claims
Elsewhere
Google Patents ↗ · USPTO patent document ↗ · USPTO Patent Center record ↗

Abstract

A CNN is applied to a histological image to identify areas of interest. The CNN classifies pixels according to relevance classes including one or more classes indicating levels of interest and at least one class indicating lack of interest. The CNN is trained on a training data set including data which has recorded how pathologists have interacted with visualizations of histological images. In the trained CNN, the in-terest-based pixel classification is used to generate a segmentation mask that defines areas of interest. The mask can be used to indicate where in an image clinically relevant features may be located. Further, it can be used to guide variable data compression of the histological image. Moreover, it can be used to control loading of image data in either a client-server model or within a memory cache policy. Furthermore, a histological image of a tissue sample of a tissue type that has been treated with a test compound is image processed in order to detect areas where toxic reactions to the test compound may have occurred. An autoencoder is trained with a training data set comprising histological images of tissue samples which are of the given tissue type, but which have not been treated with the test compound. The trained autoencoder is applied to detect tissue areas by their deviation from the normal variation seen in that tissue type as learnt by the training process, and so build up a toxicity map of the image. The toxicity map can then be used to direct a toxicological pathologist to examine the areas identified by the autoencoder as lying outside the normal range of heterogeneity for the tissue type. This makes the pathologist's review quicker and more reliable. The toxicity map can also be overlayed with the segmentation mask indicating areas of interest. When an area of interest and an area identified as lying outside the normal range of heterogeneity for the tissue type, and increased confidence score is applied to the overlapping area.

Source: PVGPATDIS, file of 2026-04-10 (ipg240917.xml).

Assignees

AssigneeTypePlaceAlso on furp.orgSource
Leica Biosystems Imaging, Inc. 131 patentsUS company or organizationVista, CAPVGPATDIS, file of 2026-04-10

Inventors

InventorPlacePatentsPerson pageSource
Walter GeorgescuVista, CA7PVGPATDIS, file of 2026-04-10
Kiran SaligramaVista, CA16PVGPATDIS, file of 2026-04-10
Allen OlsonVista, CA49PVGPATDIS, file of 2026-04-10
Girish Mallya UdupiVista, CA1PVGPATDIS, file of 2026-04-10
Bruno OliveiraVista, CA1PVGPATDIS, file of 2026-04-10

CPC categories

SubclassCategoryFirst groupSource
G06VImage or video recognition or understandingG06V10/26PVGPATDIS, file of 2026-04-10
G06TImage data processing or generation, in generalG06T7/0012PVGPATDIS, file of 2026-04-10
G16HHealthcare informatics, i.e. information and communication technology [ict] specially adapted for the handling or processing of medical or healthcare dataG16H80/00PVGPATDIS, file of 2026-04-10

Cooperative Patent Classification, as currently assigned (one row per subclass).

Assignments and security interests

RecordedWhatFromToReel/frameSource
2024-08-23Assignment
Assignment of assignor's interest
OLSON, ALLEN signed 2023-07-07
SALIGRAMA, KIRAN signed 2023-07-10
UDUPI, GIRISH MALLYA signed 2023-07-21
GEORGESCU, WALTER signed 2023-07-23
OLIVEIRA, BRUNO signed 2023-10-08
LEICA BIOSYSTEMS IMAGING, INC. VISTA, CA68382/461ECORSEXC, file of 2026-09-25
2024-01-31Assignment
Assignment of assignor's interest
OLSON, ALLEN signed 2023-07-07
SALIGRAMA, KIRAN signed 2023-07-10
MALLYA UDUPI, GIRISH signed 2023-07-21
GEORGESCU, WALTER signed 2023-07-23
OLIVEIRA, BRUNO signed 2023-10-08
LEICA BIOSYSTEMS IMAGING, INC. VISTA, CA66309/487ECORSEXC, file of 2026-09-25

Documents recorded with the USPTO Assignment Recordation Branch that list this patent. A security interest is a lender’s claim on the patent as collateral; a release ends it.

USPTO bulk data: PatentsView granted patents, disambiguated (PVGPATDIS) for patents, assignees, inventors, places and CPC categories; the USPTO Patent Assignment Dataset (ECORSEXC) for assignments and security interests. Patents filed in the last 20 years. No claims or full text. Links to other records are by name (and state): EXACT means the identical full name and state; FUZZY means a looser name match, with the confidence shown.